99.91%

Accuracy

95%

Recall

0.8636

F1 Score

0.9998

ROC-AUC

Best Model Selected: XGBoost

XGBoost was selected as the best-performing model because it achieved the highest F1 score, excellent recall, strong ROC-AUC score, and very low overfitting gap.

0.0009

Overfitting Gap

Model Comparison Table

Performance comparison of all trained models

Model	Accuracy	Precision	Recall	F1 Score	ROC-AUC	Status
XGBoost	0.999060	0.791667	0.9500	0.863636	0.999804	Best Model
Random Forest	0.999060	0.818182	0.9000	0.857143	0.999741	Strong
Extra Trees	0.998747	0.800000	0.8000	0.800000	0.999827	Good
Decision Tree	0.998433	0.727273	0.8000	0.761905	0.899529	Good
Gradient Boosting	0.997650	0.571429	1.0000	0.727273	0.999788	High Recall
KNN	0.997650	0.575758	0.9500	0.716981	0.999493	Average
AdaBoost	0.994517	0.358491	0.9500	0.520548	0.999120	Weak Precision
Naive Bayes	0.982297	0.139535	0.9000	0.241611	0.986445	Weak
Logistic Regression	0.980887	0.135714	0.9500	0.237500	0.997234	Baseline

Cross Validation

Mean CV F1 Score: 0.999725

Standard Deviation: 0.000169

This shows that the model performs consistently across different training folds.

Overfitting Analysis

Training Accuracy: 0.9999

Testing Accuracy: 0.9991

Overfitting Gap: 0.0009

The very small gap indicates that the model generalizes well and is not overfitted.

Model Performance